Description

k-mean clustering is a method of vector quantization, originally from signal processing, that is popular for cluster analysis in data mining. k-mean clustering aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean, serving as a prototype of the cluster.

(https://en.wikipedia.org/wiki/K-means_clustering)

Parameters

Name Description Type Required？ Default Value
predictionDistanceCol Column name of prediction. String
distanceType Distance type for clustering, support EUCLIDEAN and COSINE. String “EUCLIDEAN”
vectorCol Name of a vector column String ✓
maxIter Maximum iterations, the default value is 20 Integer 20
initMode Methods to get initial centers, support K_MEANS_PARALLEL and RANDOM! String “K_MEANS_PARALLEL”
initSteps When initMode is K_MEANS_PARALLEL, it defines the steps of iteration. The default value is 2. Integer 2
k Number of clusters. Integer 2
epsilon When the distance between two rounds of centers is lower than epsilon, we consider the algorithm converges! Double 1.0E-4
predictionCol Column name of prediction. String ✓
predictionDetailCol Column name of prediction result, it will include detailed info. String
reservedCols Names of the columns to be retained in the output table String[] null

Script Example

Code

  1. import numpy as np
  2. import pandas as pd
  3. data = np.array([
  4.     [0, "0 0 0"],
  5.     [1, "0.1,0.1,0.1"],
  6.     [2, "0.2,0.2,0.2"],
  7.     [3, "9 9 9"],
  8.     [4, "9.1 9.1 9.1"],
  9.     [5, "9.2 9.2 9.2"]
  10. ])
  11. df = pd.DataFrame({"id": data[:, 0], "vec": data[:, 1]})
  12. inOp = BatchOperator.fromDataframe(df, schemaStr='id int, vec string')
  13. kmeans = KMeans().setVectorCol("vec").setK(2).setPredictionCol("pred")
  14. kmeans.fit(inOp).transform(inOp).collectToDataframe()

Results

Prediction
  1. rowID   id          vec  pred
  2. 0   0        0 0 0     1
  3. 1   1  0.1,0.1,0.1     1
  4. 2   2  0.2,0.2,0.2     1
  5. 3   3        9 9 9     0
  6. 4   4  9.1 9.1 9.1     0
  7. 5   5  9.2 9.2 9.2     0